spyro at f2s.com
Thu Nov 14 15:28:34 PST 2002
On Thu, 14 Nov 2002 21:03:27 +0000 (UTC)
duncan at dwebb.ch ("Duncan Webb") wrote:
> IDE drive performance depends to a large extent on how it is connected
> to your IDE controller.
> If you want the best performance from an IDE drive, which is running
> as second hard disk then connect it to the secondary/master with
> nothing connected to your secondary slave.
Thats questionable. it might have held true on your machine, but it
makes no sense and I havent seen it on any hardware I own.
besides, bandwidth to / from the drives cache is 46MB/sec, and its the
primary master, all on its lonesome... so the bus aint the problem
> Your CD-ROM should then be connected to your primary/slave
No. this isnt right unless your controller is B0rk3n.
The only /real/ difference between a master and slave drive is the delay
in its response to probing. the slave responds (IIRC) 400ms slower than
the master. this is done so the drives always appear (initialise) in the
Other than that slave and master behave identically, and the primary /
secondary busses are entirely seperate on all good controllers (indeed
all controllers I've seen in the alst 3 years).
> and surprisingly your boot/system drive connected to the
Yes, but not for performance reasons other than avoiding probing
> Hope that's clear. I did some
> testing and the results were surprising, the same hard disk when
> connected first to the primary/master and later to the secondary/slave
> performed less that half as well when on the primary/master. You could
> try this.
Well, as I said, its the pri/master anyhow, as was its (faulty but fast)
'identical' predecessor, and also the original 20GB drive.
Thanks for the input though.
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe blfs-support' in the subject header of the message
More information about the blfs-support